About the Lab

The Multi-Modal Multi-Media Vision Research Lab (Fourm Vision Lab) conducts research that is grounded in computer vision, with a focus on publishing in top-ranked computer vision venues and working closely with domain experts to solve global/scientific challenges (medical imaging, sustainable development, conservation, astrophysics, etc.). Most of our projects involve some aspects of machine learning, image/signal processing, linear/nonlinear optimization, and large-scale data processing.
We are always interested in hard-working students (upper-level undergraduates and Master's students) with strong technical abilities, a desire to grow, and an eagerness to work on challenging problems as part of a team. While many of our members have computer science (CS) backgrounds, we encourage people with strong computational and quantitative skills from other disciplines to apply. This includes, but is not limited to, Electrical Engineering, Imaging Science, Physics, Mathematics, Statistics, and Data Science.
We work on a variety of computer-vision research areas, including:
  • remote sensing
  • medical imaging
  • multimodal learning
  • image sythesis and generative modeling
  • weakly supervised and self-supervised learning

Opening Projects

  • Generative Models Bias Mitigation
  • Fine-Grade Road Safety Estimation
  • Foundational Model Calibration
  • Cross-Domain Data Understanding

If none of these make sense for you but you think you would be a good fit for the lab in another project, feel free to email the lab director by clicking the "email" icon on the bottom of the page. Please include your CV, an explanation of the the type of projects you are interested in, and anything else you think would be useful.
Note: I have allocated all my funds at this moment. Thus, I may only consider taking thesis students at this moment.  

Qualifications

  • [Required] Non-Technical Requirement 
    • Strong work ethic and dedication
    • Intrinsic motivation in enhancing the trustworthiness of neural network-powered AI
    • Strong English oral and written communications skills
    • Thesis-track student
  • [Required] Technical Skills
    • Strong programming skills & good understanding of data structure
    • Strong mathematics foundations
    • Foundational knowledge in supervised machine learning
    • Be able to create and train simple neural networks from scratch
  • Desired but not Required Skills
    • Experience in computer vision, NLP, CNN, ViT, Transformer, and/or Diffusion model
    • Experience in research and/or technical writing

Application Instructions

  • Please send the following information to Dr. Liang by clicking the "email" icon on the bottom of the page, with the title “Interested in Trustworthy Neural Network Research”:
    • Resume/CV
    • A brief summary of background
    • Unofficial transcript
    • Indicating whether you are a current TAMU-SA student